General Performance on VicunaEval
96.3WinrateVanilla
Evaluation results
| Method | Links | ||
|---|---|---|---|
| VanillaModel=ChatGPT2023.11 | 96.3 | 37.9 | |
| VanillaModel=GPT-42023.11 | 96.3 | 38.1 | |
| Self-ReminderModel=GPT-42023.11 | 93.8 | 36.9 | |
| Goal PrioritizationModel=GPT-42023.11 | 92.5 | 36.8 | |
| VanillaModel=Vicuna-33B2023.11 | 92.5 | 36.8 | |
| Self-ReminderModel=Vicuna-33B2023.11 | 91.3 | 33.2 | |
| VanillaModel=Llama2-13B-Chat2023.11 | 91.3 | 33.8 | |
| Self-ReminderModel=ChatGPT2023.11 | 90 | 35.7 | |
| Goal PrioritizationModel=ChatGPT2023.11 | 87.5 | 34.8 | |
| VanillaModel=Llama2-7B-Chat2023.11 | 83.8 | 33.9 | |
| Goal PrioritizationModel=Vicuna-33B2023.11 | 82.5 | 33.8 | |
| VanillaModel=Vicuna-13B2023.11 | 80 | 31.9 | |
| Self-ReminderModel=Llama2-13B-Chat2023.11 | 78.8 | 30.3 | |
| Self-ReminderModel=Llama2-7B-Chat2023.11 | 77.5 | 30.3 | |
| Self-ReminderModel=Vicuna-7B2023.11 | 73.8 | 29.1 | |
| Goal PrioritizationModel=Llama2-13B-Chat2023.11 | 73.8 | 29.6 | |
| VanillaModel=Vicuna-7B2023.11 | 72.5 | 31.9 | |
| Goal PrioritizationModel=Vicuna-13B2023.11 | 72.5 | 31.1 | |
| Self-ReminderModel=Vicuna-13B2023.11 | 66.3 | 30.4 | |
| Goal PrioritizationModel=Llama2-7B-Chat2023.11 | 63.8 | 28.4 | |
| Goal PrioritizationModel=Vicuna-7B2023.11 | 50 | 26.6 |